Morphological-Rank-Linear Time-lag Added Evolutionary Forecasting method for financial time series forecasting

This paper proposes the Morphological-Rank-Linear Time-lag Added Evolutionary Forecasting (MRLTAEF) method for financial time series forecasting, which performs an evolutionary search for the minimum number of relevant time lags necessary to efficiently represent complex time series. It consists of an intelligent hybrid model composed of a Morphological-Rank-Linear (MRL) filter combined with a Modified Genetic Algorithm (MGA) which employs optimal genetic operators in order to accelerate its search convergence. The MGA searches for the particular time lags capable of a fine tuned characterization of the time series and estimates the initial (sub-optimal) parameters of the MRL filter - the mixing parameter (lambda), the rank (r), the coefficients of the linear Finite Impulse Response (FIR) filter (b) and the coefficients of the Morphological-Rank (MR) filter (a). Thus, each individual of the MGA population is trained by the averaged Least Mean Squares (LMS) algorithm to further improve the parameters of the MRL filter supplied by the MGA. Initially, the proposed MRLTAEF method chooses the most tuned prediction model for time series representation, thus it performs a behavioral statistical test in the attempt to adjust forecasting time phase distortions that appear in financial time series. Experiments are conducted with the proposed MRLTAEF method using three real world financial time series according to a group of relevant performance metrics and the results are compared to MultiLayer Perceptron (MLP) networks, MRL filters and the previously introduced Time-delay Added Evolutionary Forecasting (TAEF) method.